The objective of the proposed studies of dysarthria is two-fold: (1) to develop procedures for the quantitative assessment of neurologically impaired speakers in relation to the speaker's intelligibility and to perceived dimensions of abnormality, and (2) to identify sets or combinations of acoustic features that are characteristic of intelligibility deficits in these populations. The experimental protocol is designed to satisfy three criteria: (1) categorization of tasks in a manner that parallels tasks routinely used in neurologic examinations, (2) acquisition of data to address specific acoustic-phonetic features that are presumed to be the primary determinants of intelligibility, and (3) analysis of a sufficient range of speech behaviors to represent the various dimensions of impairment in the motor speech disorders. The primary methods of research are intelligibility judgments and perceptual ratings (to be made by a trained listener panel), acoustic analysis by spectrography and digital methods, electroglottography to monitor adductory-abductory movements of the larynx, and respiratory monitoring by means of a Respitrace system to determine the respiratory support for speech functions. Subject selection is based first and foremost on the presence of an intelligibility deficit, secondly on representativeness of large clinical populations, and thirdly on specific speech symptoms and neurologic diagnosis. We plan to study four major groups of dysarthric individuals: those with damage to either the upper motor neuron or lower motor neuron secondary to cerebrovascular lesion ("stroke" patients), persons with Parkinsons's disease, cerebral-palsied adults and children, and ALS patients. The data from these studies should contribute to improved clinical procedures for the assessment of speech intelligibility in persons with motor speech disorders.